A 22 mm torch has been developed and characterized for use in inductively coupled plasma atomic emission spectrometry. A stability curve (radio-frequency power vs. outer-gas flow rate) was constructed for the modified torch which indicates that the larger torch can be operated at flow rates and powers similar to those for an 18 mm torch. Four operating parameters were optimized by means of a simplex algorithm. Several criteria for optimization were used, including net signal intensity for Ca(II); signal-to-background noise (S/NB) for Ca(II), Mg(II), and Fe(II); and the Mg(II)/Mg(I) line-intensity ratio. The results from these simplex optimizations are compared. Mg(II)/Mg(I) was used as the final criterion for optimization for both the 22 mm and the conventional 18 mm torches. Two-dimensional spatial images of the larger plasma were compared with those of a conventional plasma (18 mm) for a variety of plasma emission features. Detection limits were determined in two ways for a suite of analytes under conditions optimized for the Mg(II)/Mg(I) ratio. The 18 mm torch affords the better limits of detection by an averaged factor of 1.5 because its smaller volume gives a lower background level. Finally, nitrogen molecular-ion emission maps were collected from both the 22 mm and a conventional 18 mm plasma as an indicator of the degree of air entrainment in the plasma. Several spatial regions of the plasma have been evaluated on the basis of the local intensity of N2+ emission for their applicability for use in atomic emission and atomic mass spectrometry. The 22 mm torch shows a larger region in and around the central channel of low or zero N2+ emission and so may be better suited for sampling into a mass spectrometer.